1. Field of the Invention
This invention relates to methods for detecting specific extracellular nucleic acid in plasma or serum fractions of human or animal blood associated with neoplastic or proliferative disease. Specifically, the invention relates to detection of nucleic acid derived from mutant oncogenes or other tumor-associated DNA, and to methods of detecting and monitoring extracellular mutant oncogenes or tumor-associated DNA found in the plasma or serum fraction of blood by using rapid DNA extraction and nucleic acid amplification. In particular, the invention relates to the detection, identification, or monitoring of the existence, progression or clinical status of benign, premalignant, or malignant neoplasms in humans or other animals that contain a mutation that is associated with the neoplasm, through detection of the mutated nucleic acid of the neoplasm in plasma or serum fractions. The invention permits the detection of extracellular, tumor-associated nucleic acid in the serum or plasma of humans or other animals recognized as having a neoplastic or proliferative disease or in individuals without any prior history or diagnosis of neoplastic or proliferative disease. The invention provides the ability to detect extracellular nucleic acid derived from genetic sequences known to be associated with neoplasia, such as oncogenes, as well as genetic sequences previously unrecognized as being associated with neoplastic or proliferative disease. The invention thereby provides methods for early identification of colorectal, pancreatic, lung, breast, bladder, ovarian, lymphoma and all other malignancies carrying tumor-related mutations of DNA, and methods for monitoring cancer and other neoplastic disorders in humans and other animals.
2. Description of the Related Art
Neoplastic disease, including most particularly that collection of diseases known as cancer, are a significant part of morbidity and mortality in adults in the developed world, being surpassed only by cardiovascular disease as the primary cause of adult death. Although improvements in cancer treatment have increased survival times from diagnosis to death, success rates of cancer treatment are more closely related to early detection of neoplastic disease that enable aggressive treatment regimes to be instituted before either primary tumor expansion or metastatic growth can ensue.
Oncogenes are normal components of every human and animal cell, responsible for the production of a great number and variety of proteins that control cell proliferation, growth regulation, and cell death. Although well over one hundred oncogenes have been described to date—nearly all identified at the deoxyribonucleic acid (DNA) sequence level—it is likely that a large number of oncogenes remains to be discovered.
Genetic mutation as the result of inborn genetic errors or environmental insult have long been recognized as playing a causative role in the development of neoplastic disease. Within the last twenty years, however, the sites of such mutations have been recognized to be within oncogenes, and mutation of such oncogenes has been found to be an intrinsic and crucial component of premalignant and malignant growth in both animals and humans. When an oncogene is mutated it alters the growth or regulation of the cell through changes in the protein it encodes. If the mutation occurs in a certain region or regions of the gene, or involves a regulatory region of a gene, a growth advantage may accrue to a cell having a mutated oncogene. Many malignant tumors or cell lines derived from them have been shown to contain one or more mutated oncogenes, and it is possible that every tumor contains at least one mutant oncogene.
Mutated oncogenes are therefore markers of malignant or premalignant conditions. It is also known that other, non-oncogenic portions of the genome may be altered in the neoplastic state. Nucleic acid based assays can detect both oncogenic and non-oncogenic DNA, whether mutated or non-mutated. In particular, nucleic acid amplification methods (for example, the polymerase chain reaction) allow the detection of small numbers of mutant molecules among a background of normal ones. While alternate means of detecting small numbers of tumor cells (such as flow cytometry) have generally been limited to hematological malignancies (Dressler and Bartow, 1989, Semin. Diag. Pathol. 6: 55–82), nucleic acid amplification assays have proven both sensitive and specific in identifying malignant cells and for predicting prognosis following chemotherapy (Fey et al., 1991, Eur. J. Cancer 27: 89–94).
Various nucleic acid amplification strategies for detecting small numbers of mutant molecules in solid tumor tissue have been developed, particularly for the ras oncogene (Chen and Viola, 1991, Anal. Biochem. 195: 51–56; Kahn et al., 1991, Oncogene 6: 1079–1083; Pellegata et al., 1992, Anticancer Res. 12: 1731–1736; Stork et al., 1991, Oncogene 6: 857–862). For example, one sensitive and specific method identifies mutant ras oncogene DNA on the basis of failure to cleave a restriction site at the crucial 12th codon (Kahn et al., 1991, ibid.). Similar protocols can be applied to detect any mutated region of DNA in a neoplasm, allowing detection of other oncogene DNA or tumor-associated DNA. Since mutated DNA can be detected not only in the primary cancer but in both precursor lesions and metastatic sites (Dix et al., 1995, Diagn. Molec. Pathol. 4: 261–265; Oudejans et al., 1991, Int. J. Cancer 49: 875–879), nucleic acid amplification assays provide a means of detecting and monitoring cancer both early and late in the course of disease.
While direct analysis of tumor tissue is frequently difficult or impossible (such as in instances of occult, unrecognized disease), peripheral blood is easily accessible and amenable to nucleic acid amplification assays such as those mentioned above. Many studies use nucleic acid amplification assays to analyze the peripheral blood of patients with cancer in order to detect intracellular DNA extracted from circulating cancer cells, including one study which detected the intracellular ras oncogene from circulating pancreatic cancer cells (Tada et al., 1993, Cancer Res. 53: 2472–4). However, it must be emphasized that almost universally these studies attempt to use nucleic acid-based amplification assays to detect extracted intracellular DNA within circulating cancer cells. The assay is performed on the cellular fraction of the blood, i.e. the cell pellet or cells within whole blood, and the serum or plasma fraction is ignored or discarded prior to analysis. Since such an approach requires the presence of metastatic circulating cancer cells (for non-hematologic tumors), it is of limited clinical use in patients with early cancers, and it is not useful in the detection of non-invasive neoplasms or pre-malignant states.
It has not been generally recognized that nucleic acid amplification assays can detect tumor-associated extracellular mutated DNA, including oncogene DNA, in the plasma or serum fraction of blood. Furthermore, it has not been recognized that this can be accomplished in a clinically useful manner, i.e. rapidly within one day, or within less than 8 hours. It is known that small but significant amounts of normal DNA circulate in the blood of healthy people (Fedorov et al., 1986, Bull. Exp. Biol, Med. 102: 1190–2; Leon et al., 1977, Cancer Res. 37: 646–50), and this amount has been found to increase in cancer states (Shapiro et al., 1983, Cancer 51: 2116–20; Stroun et al., 1989, Oncology 46: 318–322). However, these studies did not employ nucleic acid amplification methods, nor did they demonstrate the presence of mutant DNA or specific oncogene DNA in peripheral blood. Thus, the DNAs detected in blood in these reports were not definitively ascribed to cancer, nor could clinical utility be realized. In addition, it had been generally presumed by those with skill in the art that circulating extracellular DNA either does not exist or would be of no clinical utility since it would be expected to be rapidly digested by plasma DNases. However, inhibitors of DNase appear to be present in the plasma of cancer patients (Leon et al., 1981, Eur. J. Cancer 17: 533–8). Furthermore, extracellular DNA may exist in proteo-lipid complexes resistant to DNase (Stroun et al., 1987, Eur. J. Cancer Clin. Oncol. 23: 707–12). In addition, DNA from tumor cells may be present in the extracellular fluid because of secretion or shedding from viable tumor in the form of proteo-lipid complexes, release of apoptotic bodies from apoptotic tumor cells, or release of free or protein-bound DNA from necrotic or lysed cancer cells. For example, shedding of phospholipid vesicles from tumor cells is well described (Barz et al., 1985, Biochim. Biophys. Acta 814: 77–84; Taylor & Blak, 1985, “Shedding of plasma membrane fragments. Neoplastic and developmental importance,” in: Steinberg (ed) The Cell Surface in Development and Cancer, Developmental Biology, Plenum Press, New York, pp. 33–57), and similar vesicles have been shown to circulate in the blood of patients with cancer (Carr et al., 1985, Cancer Res. 45: 5944–51). Furthermore, DNA has been shown to be present on the cell surface of tumor cells (Aggarwal et al., 1975, Proc. Natl. Acad. Sci. USA 72: 928–32; Juckett & Rosenberg, 1982, Cancer Res. 42: 3565–73).
Detection of a mutant oncogene in peripheral blood plasma or serum has been the subject of reports in the prior art (see, for example, Sorenson et al., 1994, Cancer Epidemiology, Biomarkers & Prevention 3: 67–71; Vasioukhin et al., 1994, Br. J. Haematol. 86: 774–9; Vasyukhin et al., 1994, “K-ras point mutations in the blood plasma DNA of patients with colorectal tumors,” in Verna & Shamoo (eds), Biotechnology Today, Ares-Serono Symposia Publications, pp. 141–150). Mutant ras oncogenes have been demonstrated in plasma or serum using polymerase chain reaction. However, the methods employed by these groups required time-consuming and technically demanding approaches to DNA extraction and are thus of limited clinical utility. Thus, methods that permit medically useful, rapid, and timely extraction and sensitive detection of extracellular tumor-associated or extracellular mutated oncogenic DNA are not known in the art.